Before we dive into today’s innovation pattern, let’s consider the current state of the US electrical grid and services. A situation causing gain for a few, and pain for more and more of us.
In 2022, the most recent data I could find, American electricity customers lost about five and a half hours to power outages. That’s several hours longer than in most of the years between 2013 and 2019.¹ In roughly the same period, U.S. utilities have invested on the order of a trillion dollars in transmission and distribution infrastructure. And rates climbed more than 30% since 2020.² Roughly one in three U.S. households now struggles to pay utility bills or is behind on payments.³
If money alone could fix the grid, it would have by now.
The problem isn’t underinvestment. It’s overlap. And the grid isn’t the only system suffering from it.
When Abundance Becomes Disease
Ecology has a near-ironclad law called competitive exclusion: two species occupying the same niche in the same environment cannot coexist indefinitely. One outcompetes the other — not because it’s better, but because even tiny adaptive advantages compound over time.
The losing species doesn’t always die. More often, it shifts. Example: way back in the day, wolves and coyotes were competing for the same terrain. They overlapped. Over time, both adapted. Today, wolves take large prey in deep forests. Coyotes hunt smaller prey at the edges. Same landscape. Resolved overlap. Complementary roles. Ecologists describe these shifts as niche partitioning — and when traits actually diverge due to competition, character displacement.⁴ It’s how nature converts competition into cooperation.
The “thing behind the thing” insight: species don’t decide to specialize. The system pressures them into it. Scarcity—not abundance—drives differentiation.
We talked about a related finding a few weeks ago too:
Back to the context of the day job: overlap accumulates when resources are plentiful. Nobody notices four teams building the same thing when budgets are flush. Nobody questions three VPs with interchangeable job descriptions during a growth year. The duplication persists because abundance masks its cost. Then the pressure arrives — disruption in its many incarnations — and the system discovers it has few complementary parts and many competing ones.
Two Ways to Resolve Overlap
Our electrification story is a work in progress. But the history of American railroads illustrates the full arc: from abundance to overlap to differentiation and cooperation.
By the 1870s, the American railroad industry had an extraordinary problem: too many companies doing the same thing. Multiple competing lines ran parallel routes between the same cities. Each built its own track, stations, freight handling, and telegraph. It looked like a boom but it was a bust in the making. Rate wars drove prices below cost. The overlap wasn’t creating value. It was destroying it.
The resolution came in two waves.
Wave one was top-down. J.P. Morgan and the financiers merged competing lines, eliminated “redundant” routes, and rationalized the system. It looked clean on paper. But it produced monopolies, not mutualism. Communities lost leverage. Workers lost options.
Our ecologist would say that’s because the niches were assigned, not earned.
Wave two was organic. Regional railroads that survived began specializing — not because someone told them to, but because the economics of overlap became unbearable. The Illinois Central became a north-south corridor specialist. Short-line railroads found niches serving agricultural regions that trunk lines couldn’t profitably reach.
And then something important happened. Former competitors became partners. The interchange system emerged — standardized track gauges, shared freight cars, coordinated schedules — so that specialized railroads could hand off cargo seamlessly. Each company did what only it could do, and they built shared infrastructure to connect the pieces.
The lesson: top-down consolidation can initiate differentiation. But organic pressure makes it stick (through mutual benefit).
The Grid’s Incomplete Third Act
The U.S. electricity grid is stuck between acts two and three of its story.
Before deregulation, vertically integrated utilities each tried to be the whole system: generation, transmission, distribution, customer service. The niches were identical. Deregulation in the 1990s forced differentiation — generation was separated from transmission and distribution. That’s niche partitioning by regulatory design, and it worked. Specialized players emerged.
But the interchange system is incomplete. Regional transmission organizations and capacity markets exist in some regions and not others. Texas, through ERCOT, operates largely as an islanded grid with only limited ties to neighboring regions, which left it with little ability to “phone a friend” during the 2021 winter storm.⁵
Now AI-driven demand (i.e. data centers) is slamming into this half-built system. PJM, the largest U.S. grid operator serving 65 million people across 13 states, recently ran a 2027–2028 capacity auction that came about six to seven gigawatts short of its reliability requirement — a first in its history.⁶ To solve it, some data center owners are seeking on-site island generation (becoming mini-Texases, essentially) to bypass the grid entirely. This isn’t evolving the ecosystem, it’s abandoning it.⁷
The World Economic Forum calls this electricity’s “internet moment” — a transition from centralized, one-directional structures to dynamic, decentralized networks.⁸ But the interchange infrastructure that would make decentralization work is still being negotiated. So data centers and others are running ahead, building their islands. (More on that in a future article.)
What This Means for Organizations
The pattern holds at every scale of organization, from local ecosystem to multi-national enterprise: overlap accumulates in abundance, pressure forces differentiation, and mutualism only emerges when shared infrastructure makes exchange possible between the newly distinct players.
So, if niche overlap is a symptom in your team or company—which it highly likely is—resist the instinct to redraw the org chart. That’s the Morgan model. It produces compliance, not cooperation.
Instead, try the sequence nature uses.
1. Make overlap visible and costly. In nature, two wolves hunting the same elk in the same clearing both go hungry. The feedback is instant — less food, right now, for you. No intermediary absorbs the cost. No quarterly report averages it out.
In organizations, the opposite is usually true. Two teams building competing products don’t feel each other’s draw. The budget comes from the same invisible pool. Customer confusion shows up in support tickets routed to a third team. The overlap is real, but the consequences are distant — buffered by shared resources, aggregated metrics, and organizational padding. Shortening the distance between the overlap and the felt consequence is the first move.
2. Create conditions for differentiation rather than prescribing it. Ask each team to articulate its unique contribution — the thing only it provides. Make that the basis for resource allocation. This isn’t a mandate. It’s a mirror.
3. Build the interchange. This is where most organizations stop too early. Once teams have differentiated, they need shared infrastructure — common standards, handoff protocols, transparent interfaces, and the boundary crossers who thrive in the inbetween — so that what one produces, another can and does use. Without this, you get ERCOT. Specialized players who can’t help each other when it counts.
A few weeks ago we explored how vital signs — present-tense health checks instead of backward-looking KPIs — can surface where systems are thriving or struggling. Niche overlap is one of the things vital signs are designed to detect. The question this week is what you do after the diagnosis.
The answer, it turns out, is the same one nature has been using for a few billion years. Not reorganization. Differentiation, followed by interchange.
Thank you for coming on this journey with me these past 9 weeks. I hope you are finding it as interesting and informative as I am. I’d love to here your reactions and extrapolations in the comments. Onward!
Here are three “how might we” questions to start with if you see niche overlap in your organization:
How might we make overlapping teams share a visible resource ceiling — so the cost of duplication is felt by the people creating it, not absorbed somewhere else in the system?
How might we route customer confusion directly back to the teams causing it — instead of filtering it through reports that smooth away the signal?
How might we create a “niche map” before funding new initiatives — so every team can name what it uniquely contributes before resources are committed?
Sources:
¹ U.S. Energy Information Administration, “U.S. electricity customers averaged five and one-half hours of power interruptions in 2022,” November 2023.
² WIRED, “Why US Power Bills Are Surging,” 2025.
³ KBIA, “Higher electricity bills driving inflation, consumer frustration,” February 2025; The Century Foundation, “Fueling Debt: How Rising Utility Costs Are Overwhelming American Families,” 2025.
⁴ Ecologists have documented niche partitioning and interference competition between wolves and coyotes across shared landscapes. See: Mastro, L.L., “Evolution in coyotes in response to the megafaunal extinctions,” PNAS, 2011; Newsome, T.M. & Ripple, W.J., “Interference competition between wolves and coyotes during variable prey abundance,” Ecology and Evolution, 2021.
⁵ National Hydropower Association, “Winter 2021 Storm Event in Texas: An Assessment of the Energy System Reliability Failures,” 2021.
⁶ Introl, “PJM’s 6GW Capacity Shortfall: The Grid Crisis That Could Reshape Energy Markets,” 2025.
⁷ Reuters, “US grid rules for faster data centers favor on-site gas plants,” January 2026.
⁸ World Economic Forum, “Energy experts on building the power systems of the future,” January 2026.
Connections to The Insider’s Guide to Innovation at Microsoft
Language as strategic tool — How you name the intervention shapes how people experience it. “Reorganization” signals top-down assignment. “Differentiation” signals agency.
Assumption mapping — Overlapping teams often exist because the original rationale has changed but the structure hasn’t. Revisiting assumptions is how you find the niches that no longer need to exist.
Double-loop learning — The practice of asking “Now that we know what we know, what is possible?” breaks conservation-phase thinking and opens the door to differentiation.
Biomimicry — The adaptive cycle: meadow → forest → release → reorganization. Niche overlap accumulates in the forest phase. Scarcity initiates the release.
Boundary Crossers — The people who manage the interchange between differentiated teams, translating across niches so that specialized groups can exchange value without losing their distinctiveness.
This piece is part of Innovating Out Loud, a weekly practice of sharing research and sense-making at early iterations rather than polished conclusions — a companion to the monthly IOL webcast. Register at regenerouslabs.com/innovatingoutloud.
AI Disclosure: This piece was researched with Perplexity and written and edited with my custom Claude writing partners. The ideas, research direction, and editorial judgment are mine. The audio is also me, flubs and all.
